Adaptive Training of Manual Control: Relation of Adaptive Scheme Parameters to Task Parameters.
Abstract
Five adaptive scheme parameters were studied in a series of adaptive training experiments designed to answer questions concerning optimal parameter values for adjustment of task difficulty. Three highly automated experiments tested 194 subjects on a multi-axis task simulating level flight. The five parameters studied were Performance Measurement Interval (PMI), turbulence Quasi-Stationary Interval (QSI), Progression:Regression ratio for task difficulty (P:R), within-trial Feedback of Difficulty Level (FBK) and vector vs. component performance indices. Subject to clarification of possible interaction between PMI and phugoid frequency, results suggested that duration for PMI should be at least 20 sec or three times the reciprocal of the phugoid natural frequency. Ten sec PMI was found to be undesirable for conditions of this study. Outcome for QSI suggested that training may be most effective with only a modest disturbance -- just sufficient to initiate subject inputs. P:R was found to be a significant factor but fixed rates of change of difficulty were shown to be inferior to rates proportional to error. FBK was found to improve only transfer task performance. (Modified author abstract)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1973
- Accession Number
- AD0764901
Entities
People
- D. A. Norman